Virtual Assistant Data Manager for Privacy-Aware Data Sharing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing virtual assistants lack effective mechanisms to validate and control data sharing requests, potentially leading to the unauthorized disclosure of sensitive user information.

Innovation Solution

Implement a data manager module within virtual assistants to evaluate data sharing requests using predefined templates, trust factors, user consent, and privacy policies, calculating an overall confidence score to determine whether to share user data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If virtual assistants share user data with requesting entities, then service functionality and user interaction capabilities are improved, but data privacy and security are compromised

Engineering Contradiction:

Improvedata sharing capabilityVSAvoiddata privacy risk

Core Design Contradiction:

Adaptability or versatilityVSObject-affected harmful factors

Solution Approach 1:

The patent introduces a data manager module as an intermediary between the virtual assistant and requesting entities. This mediator evaluates data sharing requests against privacy policies, trust factors, and contextual information before allowing data access, thus enabling controlled data sharing that balances functionality with privacy protection

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent implements dynamic data sharing control where the virtual assistant adjusts data sharing behavior based on real-time evaluation of requests, user preferences, contextual factors, and trust levels. This dynamic approach allows the system to be adaptable to different situations while maintaining privacy safeguards

Inventive Principle:

Principle #15Dynamics

2Object-affected harmful factors

If virtual assistants implement comprehensive data sharing controls, then data privacy is improved, but system complexity increases

Engineering Contradiction:

Improvedata privacy protectionVSAvoidsystem architecture complexity

Core Design Contradiction:

Object-affected harmful factorsVSDevice complexity

Solution Approach 1:

The patent divides the virtual assistant system into distinct functional modules, including a data manager module that handles privacy evaluation separately from the main assistant logic. This segmentation allows comprehensive privacy controls to be implemented in a modular fashion, reducing overall system complexity

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements preliminary evaluation of data sharing requests against privacy policies and trust factors before actual data access occurs. By performing these privacy checks in advance, the system establishes clear decision boundaries that simplify the overall control architecture

Inventive Principle:

Principle #10Preliminary action

3Measurement precision

If virtual assistants evaluate each data request individually, then data privacy accuracy is improved, but processing time increases

Engineering Contradiction:

Improvedata sharing decision accuracyVSAvoidrequest processing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent implements a tiered evaluation approach where the data manager module performs essential privacy checks on all requests, and more detailed contextual analysis only when initially required. This partial action approach maintains high decision accuracy while reducing average processing time for routine requests

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS12493806B2Artificial intelligence enabled data privacy for virtual assistants

Publication Date: 2025.12.09 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

According to one embodiment, a method, computer system, and computer program product for evaluating a data request is provided. The embodiment may include receiving a request to share data of a user during an interaction between an intelligent virtual assistant (IVA) representing the user and a requesting entity. The request is sent from the requesting entity to the IVA. The embodiment may include calculating an overall data sharing confidence score based on an evaluation of the request. The embodiment may include determining whether the overall data sharing confidence score is above a data sharing threshold value. In response to determining that the overall data sharing confidence score is above the data sharing threshold value, the embodiment may include instructing the IVA to share the data of the user with the requesting entity.